^{page 231 note 1} A list of scholars ranged on either side is given in Bidez, , Eos, ou Platon et I'Orient (1945), p. 102Google Scholar, following de Strycker.

^{page 231 note 2} C.Q.N.S. iii (1953), 46.Google Scholar

^{page 231 note 3} What Plato Said (1933), p. 415.

^{page 231 note 4} In Bidez, , op. cit., p. 121.Google Scholar

^{page 231 note 5} Origin and Growth of Plato's Logic (1897), P. 197.Google Scholar

^{page 231 note 6} Untersuchungen über Platon (1888), p. 89.Google Scholar

^{page 231 note 7} To these may be added one , one , and one which are all much more frequent in the later period.

^{page 231 note 8} The only points at all surprising are the absence of superlative answer-forms like and the low number of prepositions—11·5 to a page of Didot.

^{page 231 note 9} For this I rely on the tables in Ritter's, Untersuchungen, pp. 56–57Google Scholar, corrected from his Platon, i (1910), pp. 236–7.

^{page 232 note 1} The Clitopho has no answer-formulae.

^{page 232 note 2} On my calculations he uses 60 different short affirmative answers on 213 occasions, and 22 negative answers on 54 occasions, but uses none of these 4 affirmative and 1 negative forms. In general, he tends to use Plato's earlier forms much as Plato does himself, except that he uses many more expressions of the form , etc. The passages I have examined are: Memorabilia, i. 2. 34–48, 3. 8–4. 12, ii. 1. 1–12, and 2–10, and the whole of iii and iv, the Oeconomicus and the Symposium.

^{page 232 note 3} I have chosen 126 as a convenient but arbitrary dividing line. It is not a point on which certainty is possible, but if a division is to be made it must come hereabouts. It is interesting that Taylor, (Plato, the Man and His Work⁶ (1949), p. 525Google Scholar) notes confusion of exposition and an abrupt transition at 127 d.

^{page 232 note 4} C.Q. N.s. iii (1953), 80.Google Scholar

^{page 232 note 5} J. Philol. xxxv (1920) 225–56Google Scholar. These are markedly prominent in the Sophist digression and later dialogues. But see below.

^{page 232 note 6} For these I have used the relevant parts of Lutoslawski, , Origin and Growth of Plato's Logic, pp. 76–140.Google Scholar

^{page 232 note 7} Ritter, , Untersuchungen, p. 57.Google Scholar

^{page 232 note 8} For some examples see below. I have used also Dittenberger (corrected by Ritter) whose figures are given by Lutoslawski, , op. cit., p. 104Google Scholar, and Ritter, , Untersuchungen, p. 58Google Scholar, corrected from Bursian's, Jahresbericht 187 (1921), 48Google Scholar. The relevant points are the usages of and the consistent use of .

^{page 233 note 1} Denniston, , The Greek Particles (1934), p. 493Google Scholar, gives a number of similar cases.

^{page 233 note 2} See Lutoslawski, , op. cit., p. 123, no. 336.Google Scholar

^{page 233 note 3} Ibid., p. 119, no. 303, from Kugler.

^{page 233 note 4} It is not possible to give a tidy list. I have selected these tests from Lutoslawski, and the figures I give should be compared with those given by him and his authorities. As might be expected, some tests show no significant difference between the two parts, and many have no application to this dialogue.

^{page 233 note 5} Lutoslawski, p. 122, no. 318. Cf. Ritter, , Untersuchungen, pp. 16–17Google Scholar, and table p. 56. Plato's figures show an overall decline. Ritter gives Rep. 5·5 per cent.

^{page 233 note 6} Lutoslawski, p. 130, no. 391, from Lina. Plato in all has about 5 . to 2 . But the figures given above may be a freak result.

^{page 233 note 7} Ibid., p. 137, no. 448, from von Arnim. Plato's figures fall overall.

^{page 233 note 8} Ibid., p. 137, no. 451, from von Arnim. Plato's figures gradually increase. Rep. 1·35 per cent. Theaet. 11·5 per cent.

^{page 233 note 9} Ibid., p. 12, no. 378, from Siebeck. Plato's figures increase. Theaet.17 per cent.

^{page 233 note 10} For these see J. Philol. xxxv (1920), 225 ffGoogle Scholar

^{page 234 note 1} e.g. the weakness of characterization noted by Bruns, Das literarische Porträt der Griechen (1896), p. 339Google Scholar, the grotesque situation with which the dialogue opens, and the linguistic objections to its authenticity and the philosophical objections to an early dating, both discussed below.

^{page 234 note 2} The ratio of the two parts, based on the Didot edition, is 17:8.

^{page 234 note 3} What Plato Said, p. 652.

^{page 234 note 4} In unpublished papers, which he has been kind enough to let me see. I have learned much from them.

^{page 234 note 5} In Bidez, , Eos ou Platon et L'Orient, pp. 104, 115.Google Scholar

^{page 234 note 6} I should add a general caution that defenders of the Alcibiades, in particular Vink, (Plato's Eerstt Alcibiades (1939Google Scholar), have questioned the value of these so-called imitations, and rightly point out that there are many similarities of thought and phrasing between undoubtedly genuine dialogues. However, in this case the indebtedness to other works seems so great that I am inclined to believe that the general effect of copying cannot be discounted. Similarly the imitations of Aeschines, Antisthenes, and possibly Xenophon found by Dittmar, (Aischines von Sphettos (1912), pp. 163–77Google Scholar) should probably be accepted, though it might be better to call them reminiscences rather than imitations. Wilamowitz, too (Platon, ii (1919) 325–6Google Scholar) notes the influence of die early dialogues, but can find no trace of the effects of the later ones. He thinks also that the Theages, probably a fourth-century work, is dependent on the Alcibiades which would give that a fairly early date.

^{page 234 note 7} Vit. Alc. i.

^{page 234 note 8} 133 c.

^{page 235 note 1} Eudoxus' life is very obscure. I follow Hultsch in R.E. vi. 913–14 in supposing that the Enoptron was based on observations made at Cyzicus before he joined Plato in Athens, and the Phainomena on observations made from Athens.

^{page 235 note 2} In Bidez Eos, pp. 113–14.

^{page 235 note 3} Ibid., pp. 116–17.

^{page 235 note 4} Ibid., p. 118.

^{page 235 note 5} Ibid., pp. 120–1.

^{page 235 note 6} In unpublished papers.

^{page 235 note 7} Ibid.

^{page 235 note 8} Plato's Eerste Alcibiades, p. 138. Vink defends the work's authenticity.

^{page 235 note 9} C.Q.. N.s. iii. 46–52.

^{page 235 note 10} 129 b, 130 d.

^{page 235 note 11} Der Grosse Alcibiades, ii. 17.

^{page 235 note 12} e.g. 251 b.

^{page 236 note 1} Croiset, ((Euvres complètes de Platon, i, 1920Google Scholar) translates: ‘ce que c'est au juste que soi-même’.

^{page 236 note 2} Croiset translates: ‘il fallait chercher d'abord ce que c'est que ‘soi-même’. Or, au lieu du “soi-meme” en sa totalité, nous avons cherché ce qu'est chaque “soi-même’ en particulier.’ He takes this to mean that till now we have simply distinguished body and soul in man, but must now go on to distinguish the parts of the soul. This explanation is not a satisfactory one, but I can think of none better.

^{page 236 note 3} Frag. 46.

^{page 236 note 4} In unpublished papers.

^{page 236 note 5} e.g. Friedländer, , Platon, ii (1930), 243–4 and Croiset, (Euvres complètes de Platon, i.Google Scholar

^{page 236 note 6} Philologische Wochenschrift, lii (1932), 700–2.Google Scholar

^{page 237 note 1} On the other hand the disturbance in Stobaeus might be an indication that the text had been tampered with.

^{page 237 note 2} Rev. de Phibl. xlv (1921), 87–89.Google Scholar

^{page 237 note 3} Similar errors occur in the manuscripts, eg. at Symp. 190 e 8, 191 a 2, and 202 a 5.

^{page 237 note 4} 218 d onwards.

^{page 238 note 5} 126b–I27d.

^{page 238 note 6} C.Q..N.S. iii (1953), 50.

^{page 238 note 7} It is worth noting that the Clitopho tackles the same difficulties in much the same way.

^{page 238 note 1} 132d–133c.

^{page 238 note 2} 245 b–257 a.

^{page 238 note 3} 255 d:

^{page 238 note 4} C.Q.. N.S. iii. 47.

^{page 238 note 5} 246 a–c. The relationship between the twelve astronomical gods of the Phaedrus and the single god of the Alcibiades and other dialogues is an obscure one, which I shall discuss later.

^{page 239 note 1} 47 b, c. I follow Owen, G. E. L. (C.Q.. N.s. iii (1953), 79–95CrossRefGoogle Scholar) in placing the Timaeus much earlier than has been the accepted practice. It must come after the Republic, but may be before the Theaetetus and the Phaedrus.

^{page 239 note 2} 176 b–c. Cf. Rep. 613 a. Theaet. 173c–174 b is also noteworthy, showing that for Plato at this time one of the philosopher's chief interests was astronomy.

^{page 239 note 3} For further discussion of these matters, see Dodds, E. R., ‘Plato and the Irrational’ in J.H.S. xlv (1945), 16–25Google Scholar. I note m particular his suggestion (p. 24) that Plato under went a second religious conversion under the influence of Eudoxus.

^{page 239 note 4} 133 c.

^{page 239 note 5} Bluck, (C.Q.. N.s. iii. 49Google Scholar) would like to see in (133 c) a reference to the Enoptron of Eudoxus, an astronomical work written probably before he joined Plato in Athens. The weakness of this attractive view is that there seems to be no necessity for it: is called for naturally by the context, where v cannot be repeated. And a study of the occurrences of these two words leads me to believe that was reserved almost exclusively for mirrors, i.e. objects manufactured for that purpose, while means any reflecting surface, and would therefore be the right word in the passage under discussion. (The distinction can be maintained, I believe, though it is not necessary, in the long passage of Aristotle, 459 b–460 a.) However, Bluck's other interpretation of the passage is unsatisfactory, because it would not fully complete the simile. We have begun with reflections in mirrors and in the eyes of our friends; we should end, by chiasmus, with reflections in the souls of our friends and in something much larger, i.e. God. Hence, I would prefer the astronomicainterpretation with or without Eudoxus.

^{page 240 note l} Theaet. 7 in 50, Soph. 4 in 40, Pol. 5 in 43, Phaedr. 2 in 39, Parm. 10 in 31.

^{page 240 note 2} 115 c 7, 124c9, 124e 13.

^{page 240 note 3} Above, p. 234.

^{page 240 note 4} Above, p. 233